Copper deficiency in infants fed cow milk

786

Clinical and laboratory observations

The Journal of Pediatrics May 1985

Copper deficiency in infants fed cow milk Yael Levy, M.D, Avraham Zeharia, M.D., Michael Grunebaum, M.D., Menachem Nitzan, M.D., and Reuben Steinherz, M.D. P e t a h Tiqva, israel

NUTRITIONAL COPPER DEFICIENCY in infancy has been observed by Tanaka et al. ~ in a premature baby fed milk formula low in copper. Copper deficiency has also been reported in infants given total parenteral nutrition w i t h o u t copper supplementation in cases of severe diarrhea, malabsorption, protein loss, or the use of chelating agents. 2-5 Copper deficiency m a y cause anemia and other subtle symptoms that may go undiagnosed but resolve with diet enrichment. We describe two infants with copper deficiency, and draw attention to the similarities between our patients and those of Tanaka. CASE REPORTS

Patient 1. This 6-month-old Arab boy was hospitalized because of vomiting and apathy. He had been born after an uneventful pregnancy and delivery; birth weight was 3500 gm. Since birth he had been fed only cow milk. On physical examination he was pale but well nourished. (weight 8600 gm, length 73 cm). The child's symptoms were attributed to aseptic meningitis. Laboratory tests revealed hypochromic microcytic anemia (hemoglobin 8 gm/dl, with reticulocyte count 0.1%). The leukocyte count was 2700/mm 3, with 8% neutrophils. Serum iron concentration was 16 #g/dl. Serum vitamin B12 and folic acid levels were within the normal range. Glucose-6-phosphate dehydrogenase level and hemoglobin electrophoresis were normal. Bone marrow aspiration disclosed vacuolated metamyelocytes and normoblasts, with maturation arrest of the myeloid line but no abnormal cells. Serum copper concentration in serial examinations ranged between 11 and 30 #g/dl (normal 85 to 110 t~g/dl). Ceruloplasmin levels varied between 11 and 25 OD units (normal range 280 to 570 OD units). The infant's diet was enriched with meat, vegetables, and fruits and supplemented with iron. Four months later normal levels of copper and ceruloplasmin were reached (130 #g/dl and 663 OD units, respectively), with hemoglobin 10.1 gm/dl and leukocyte count 9000/mm 3 (Fig. l). Patient 2. This 6-month-old infant boy of Jewish Ashkenazi origin was hospitalized because of fever and mild respiratory From the Departments of Pediatrics .4 and Pediatric Radiology and Section on Human Biochemical and Developmental Genetics, Beilinson Medical Center, Petah Tiqva; and Tel-Aviv University Sackler School of Medicine. Submitted for publication July 13, 1984; accepted Oct. 8, 1984. Reprint requests: Reuben Steinherz, M.D., Department of Pediatrics A, Beilinson Medical Center, Petah Tiqva 49 100, Israel.

distress. He was born after an uneventful pregnancy and labor; birth weight was 2900 gm. His diet consisted mainly of cow milk. Physical examination revealed an infant in good nutritional condition; his weight of 5100 gm was below the 3rd percentile for age, but his rate of growth was satisfactory. The infant had physical findings of ventricular septal defect and bilateral otitis media. Joint laxity and synovial clicks prompted us to perform radiographic examination of the hip joints, which revealed increased density of the preparatory calcification areas with spur formation at the proximal parts of the femurs (Fig. 2). These radiologic findings were compatible with either ascorbic acid deficiency or copper deficiency. Vitamin C deficiency was excluded by normal results of vitamin C tolerance test. Serum copper concentration ranged between 11 and 27 ~g/dl, and cerulopiasmin levels were 9 to 25 OD units. The hemoglobin level was 9 gm/dl, the retieulocyte count 4.2%, and the leukocyte count 9900/ram 3, with only 17% neutrophils. Serum iron concentration was 37 #g/dl, and ferritin level was normal at 18 ng/ml. Three months after admission and diet enrichment with meat and vegetables, normal levels of copper and Ceruloplasmin were achieved (149 ~g/dl and 349 OD units, respectively). Hemoglobin was stabilized at 12.4 gm/dl, Radiographs of the long bones at that time revealed complete recovery. DISCUSSION The manifestations of copper deficiency have been extensively detailed in previous reports? ,6 The clinical signs include hypotonia, pallor, hypopigmentation, hepatosplenomegaly, prominent scalp veins, and psychomotor retardation. The laboratory findings include neutropenia and iron therapy-resistant hypochromic anemia, and 10w levels of serum copper and ceruloplasmin. 6,7 Radiologic changes in long bones include osteoporosis, metaphyseal cupping, and spur formation. 8 Bone marrow changes include vacuolated erythroid and myeloid cells and maturation arrest of the myeloid line. 7 Our two patients probably had unusual examples of copper deficiency: neither was premature, neither had any gastrointestinal tract disorder, and both appeared well nourished. The clue to copper deficiency in patient 1 was neutropenia with granulocytopenia. The leading sign of copper deficiency in patient 2 was the spur formation in the rnetaphyseal region o f the two femurs (the x-ray survey of the pelvis was intended to exclude hip dislocation, not to prove copper deficiency).

Volume 106 Number 5

Clinical and laboratory observations

78 7

Hemoglobin ~1tq1--. . . . WBC/mrr Hb.r/d,

Norma4range o~ c o r ~ , r ~ j-41

I(X0OO

150 ~'140

8OO0

-120 -110 -100 -90 -80 -70 -60 "50 -40 -30 -20 -10

7O00 6O00

9

I

iiiiii

i:i:ii, i/.,~//(/

ii:

500O 4000 8

2000.

.~,~ " \,\ w . / , / , J "

3000 7

i.

I

~

/II II

10OO

=

~

:i !i!

Age (months) Iron Supplementation

Fig. 1. White blood cell counts and hemoglobin levels in correlation with serum copper levels. The two infants had hypochromic microcytic anemia. In patient 2 the ferritin level was high, indicating enough iron in the body stores. Patient 1 was given iron supplementation. But in both patients, anemia was resolved only when copper levels returned to normal. A characteristic feature of our two patients, as in the previously reported patients of Naveh et al. 9 and Tanaka et al., ~is that since birth the infants' diet was based solely on cow milk. The recommended dietary copper allowance is 80 #g/kg/day for a full-term infant, ~~and the recommended minimum concentration of copper in infant formulas is 0.42 #g/ml. ]~ Cow milk is a poor source of copper (0.1 +_ 0.2 gg/ml)? 2 Human milk is richer in copper (0.27 _+ 0.02 ug/ml), and its bioavailability is higher than in cow milk; a higher proportion of copper is bound to the casein in cow milkJ 2 In contrast to the patients of Naveh et al. and Tanaka et al., our two infants did not receive any copper supplementation; the only change in their diet was the addition of chicken, meat, and vegetables. The decision to manage with dietary manipulation without copper supplementation was based first on the relatively minimal clinical changes observed in the ehildren, which could easily have been undiagnosed or misdiagnosed. (We did not believe that the otitis media or the aseptic meningitis could be attributed to impaired immunity secondary to the leukopenia or neutropenia, because both are common reasons for medical attention at this age.) Second, based on Tanaka's patient and others, in addition to our own experience with patients with severe copper deficiency, reticulocytosis starts quite early on institution of copper therapy but hemoglobin reaches normal values only after a few months of therapy. Thus, there is no need for copper supplementation in mild

Fig. 2. Pelvic radiograph demonstrating spur formation (arrows) on both femurs.

cases. Finally, the copper content in meat, and especially in poultry and vegetables, is high (2 to 9 #g/ml). ]3 It seemed to us that dietary changes could replace copper supplementation and probably improve copper balance in these particular infants. We believe that copper deficiency in the first year of life in infants fed cow milk is much more common than is thought, and may go undiagnosed. Copper deficiency in otherwise normal infants can develop at around the age of 6 months, probablyqSecause of exhaustion of the liver copper stores, a diet based on cow milk only, and a significant increase in body weight. This ffaechanism is somewhat similar to the appearance of nutritional iron deficiency anemia observed at the same age. This copper

788

Clinical and laboratory observations

deficiency is transient; dietary changes normally resolve both iron and copper deficiency. The recovery from nutritional anemia in the first year of life is usually attributed to the more prevailing iron deficiency. The possibility of copper deficiency or inability to utilize iron because of copper deficiency is masked because it also can be resolved by dietary changes.

REFERENCES 1. Tanaka Y, Hatano S, Nishi Y, Usui T: Nutritional copper deficiency in a Japanese infant on formula. J PEDIATR52:255, 1980. 2. Ashkenazi A, Levin S, Djaldetti M, et al: The syndrome of neonatal copper deficiency. Pediatrics 52:525, 1973. 3. Karpel JT, Peden VH: Copper deficiency in long-term parenteral nutrition. J PEDIATR 80:32, 1972. 4. Heller RM, Kirchner SG, O'Neill JA Jr, et al: Skeletal changes of copper deficiency in infants receiving prolonged total parenteral nutrition. J PEDIATR92:947, 1978. 5. Williams DM: Copper deficiency in humans. Semin Hematol 20:118, 1983.

The Journal of Pediatrics May 1985

6. Shaw JCL: Trace elements in the fetus and young infant, lI. Copper, manganese, selenium, and chromium. Am J Dis Child 134:74, 1980. 7. Dunlop WM, James GW, Hume DM: Anemia and neutropenia caused by copper deficiency. Ann Intern Med 80:470, 1974. 8. Grunebaum M, Horodniceanu C, Steinherz R: Radiographic manifestations of bone changes in copper deficiency. Pediatr Radiol 9:101, 1980. 9. Naveh Y, Hazani A, Berant M: Copper deficiency with cow's milk diet. Pediatrics 68:397, 1981. 10. Walravens PA: Nutritional importance of copper and zinc in neonates and infants. Clin Chem 26:185, 1980. 11. Barness LA: Nutritional requirements of the full-term neonate. In Suskind RM, editors: Textbook of pediatric nutrition. New York, 1981, Raven Press, p 21. 12. Fransson GB, Lonnerdal B: Distribution of trace elements and minerals in human and cow's milk. Pediatr Res 17:912, 1983. 13. Delves HT: Dietary sources of copper. In Biological roles of copper. CIBA Foundation Symposium 79 (new series). Amsterdam, 1980, Excerpta Medica, p 6.

Epidural spinal cord compression as the initial finding in childhood acute leukemia and non-Hodgkin lymphoma Ching-Hon Pui, M.D., Gary V. Dahl, M.D., H. Omar Hustu, M.D., and Sharon B. Murphy, M.D. Memphis,

Tennessee

EPIDURAL spinal cord compression is a rare but serious complication in patients with leukemia ~-3 or non-Hodgkin lymphoma. 4-9 Prompt diagnosis and treatment are necessary if permanent neurological sequelae are to be avoided. Most reported cases have occurred in adults with relapsed N H L 6, 7; hence, very little is known about this complication in children, particularly those with leukemia. W e summarize the clinical characteristics and treatment responses of 11 children with epidural cord compression at diagnosis of acute leukemia or N H L .

From the Leukemia-Lymphoma Division and Radiation Therapy Division, St. Jude Children's Research Hospital. Submitted for publication Sept. 4, 1984; accepted Oct. 8, 1984. Supported in part by Grants CA-20180 and CA-21765 from the National Cancer Institute and by the American Lebanese Syrian Associated Charities. Reprint requests: Ching-Hon Pui, M.D., St. Jude Children's Research Hospital, 332 North Lauderdale, P.O. Box 318, Memphis, TAT 38101.

METHODS From 1962 to 1984, 1412 children with acute lymphoblastic leukemia, 309 with acute nonlymphoblastic leukemia, and 276 with N H L were admitted to St. Jude Children's Research Hospital. A t diagnosis, seven children with acute leukemia and f o u r with N H L were fouhd to have epidural spinal cord compression resulting from their NHL ALL ANLL

Non-Hodgkin lymphoma Acute lymphoblastic leukemia Acute nonlymphoblastic leukemia

malignant diseases. The diagnoses and levels of spinal cord compression were substantiated by iophendylate or metrizamide myelography in 10 of the 11 patients; two also underwent cisternal myelography. In four patients, the location of lesions was confirmed by laminectomy, done before r J e r r a l . Spine roentgenography, as well as examinations of cerebrospinal fluid and bone marrow, was performed in all patients. Lymphoma was diagnosed from